A method for estimating the mass of a pure electric vehicle under all-conditions considering cornering conditions

Abstract

This patent discloses a method for estimating the mass of a pure electric vehicle under all working conditions, which takes into account the turning condition. The method establishes a mass estimation model of a recursive least square method with a forgetting factor under a straight-line condition through the longitudinal dynamics model of the vehicle. On this basis, a turning resistance term is introduced into the straight-line condition estimator to establish a mass estimation model under all working conditions. This method includes the following steps: 1. Based on the external force of the vehicle and the vehicle driving balance equation, a longitudinal dynamics model of the pure electric vehicle is established; 2. Based on the longitudinal dynamics model, a mass estimation algorithm of a recursive least square method with a forgetting factor under a straight-line condition is established; 3. Based on the longitudinal force analysis of the vehicle under the turning condition, a turning resistance term is introduced into the straight-line condition estimator to establish a mass estimation model under all working conditions; This method improves the estimation effect of the vehicle mass estimation under the turning condition by analyzing the longitudinal force of the turning condition and introducing the turning resistance term into the mass estimator.

Description

Technical Field

[0001] The present invention belongs to the field of vehicle dynamics, and in particular relates to a method for estimating the mass of a pure electric vehicle under all-operating conditions taking into account a turning condition. Background Art

[0002] As society pays more attention to new energy technologies, new energy vehicles are gradually emerging and are favored by consumers. As a type of new energy vehicle, the vehicle mass of pure electric vehicles is an important parameter of vehicle longitudinal dynamics. For active safety control of vehicles, especially when controlling the expected speed and acceleration of the vehicle output, it is necessary to know more accurate vehicle mass parameters.

[0003] Most of the current mass estimation methods use the longitudinal dynamics equation of the vehicle under straight-line conditions to perform mass estimation. However, the vehicle may encounter turning, lane changing and other conditions during driving. During turning conditions, the mass estimator designed based on the longitudinal dynamics equation of the vehicle under straight-line driving conditions can no longer accurately estimate the mass of the entire vehicle.

[0004] Existing patents, such as the invention patent with patent number CN 118651238 A, use the vehicle terminal to access the on-board diagnostic system OBD to obtain the required vehicle operation data, and obtain a continuous uniform speed or acceleration data segment greater than n+1 sampling cycles; based on the data segment, use the principle of conservation of energy to iteratively calculate the vehicle mass and other related parameters within an acceleration or deceleration segment; continuously update the iterative results, and finally store and output the results.

[0005] This method relies heavily on the accuracy of the vehicle operating data obtained by the on-board diagnostic system (OBD). However, the vehicle status information data is interfered by noise and electromagnetic signals. At the same time, the vehicle status information processed by filtering may cause problems such as vehicle information distortion. Summary of the invention

[0006] This patent discloses a method for estimating the mass of a pure electric vehicle under all working conditions, which takes into account the turning condition. The method establishes a mass estimation model of a recursive least square method with a forgetting factor under a straight-line condition through the longitudinal dynamics model of the vehicle. On this basis, a turning resistance term is introduced into the straight-line condition estimator to establish a mass estimation model under all working conditions. This method includes the following steps: 1. Based on the external force of the vehicle and the vehicle driving balance equation, a longitudinal dynamics model of the pure electric vehicle is established; 2. Based on the longitudinal dynamics model, a mass estimation algorithm of a recursive least square method with a forgetting factor under a straight-line condition is established; 3. Based on the longitudinal force analysis of the vehicle under the turning condition, a turning resistance term is introduced into the straight-line condition estimator to establish a mass estimation model under all working conditions; This method improves the estimation effect of the vehicle mass estimation under the turning condition by analyzing the longitudinal force of the turning condition and introducing the turning resistance term into the mass estimator.

[0007] In order to solve the above technical problems, the present invention is implemented by adopting the following technical solutions:

[0008] A method for estimating the mass of a pure electric vehicle in all working conditions considering a turning condition, characterized by comprising the following steps:

[0009] S1, based on the vehicle driving external force and vehicle driving balance equation, establish the longitudinal dynamics model of pure electric vehicle;

[0010] S2, based on the longitudinal dynamics model, a mass estimation algorithm of the recursive least square method with forgetting factor under straight-line working conditions is established;

[0011] S3, based on the longitudinal force analysis of the vehicle under turning conditions, the turning resistance term is introduced into the straight-line condition estimator to establish a full-condition mass estimation model;

[0012] The specific process of step S1 is as follows:

[0013] When a car accelerates on a slope, the resistance and the driving force of the car are balanced. The car driving equation is as follows:

[0014] F t =F f +F w +F i +F j (1)

[0015] In the formula, F t is the driving force, F f is the rolling resistance, F w is the air resistance, F i is the slope resistance, F j For acceleration resistance;

[0016] The longitudinal dynamic model of the car during driving is:

[0017]

[0018] In the formula, F t is the longitudinal driving force of the car, m is the mass of the car, ρ is the air density, C d is the air resistance coefficient, θ is the road slope, and f is the rolling resistance coefficient;

[0019] Accelerate the longitudinal dynamics model of the vehicle in the above formula Differentiating it gives:

[0020]

[0021] Considering that the road slope changes little, the influence of the road slope on the acceleration is small during the driving process of the car, so the result of the slope on the acceleration can be approximated to zero and can be ignored, that is:

[0022]

[0023] The driving torque of pure electric vehicles changes very frequently. When the driving torque of the motor changes frequently, the rate of change of the vehicle acceleration is higher, while the rate of change of the vehicle speed is much smaller than the rate of change of the acceleration. Therefore, the derivative of the rate of change of the acceleration with respect to the rate of change of the speed is approximately zero and can be ignored, that is:

[0024]

[0025] The signal collected by the sensor is filtered. Since the same filter is used, according to the filter principle, combined with equations (4, (5)), equation (3) can be transformed into:

[0026]

[0027] In the formula, F * x is the driving force signal, v * x is the acceleration;

[0028] The specific process of step S2 is as follows:

[0029] Using Newton's second law to analyze the forces during the driving process of a car is:

[0030] F=ma (7)

[0031] In the formula, F is the resultant force on the car during driving, m is the mass of the car, and a is the acceleration of the car.

[0032] The only parameters that can be transmitted from the vehicle controller via CANape are the output shaft speed and the actual output torque. When estimating the vehicle mass, the effects of air resistance, rolling resistance and road slope are ignored. The resultant force and acceleration of the vehicle required by Newton's second law can be calculated from the actual output torque and output shaft speed of the drive motor through the vehicle dynamics equation. The equation for calculating the driving force is:

[0033]

[0034] In the formula, F t For the driving force of the car, T tq is the actual output torque of the driving motor, i gis the main reduction ratio of the automatic transmission, i0 is the transmission ratio of the automatic transmission, η is the mechanical efficiency of the transmission system, and r is the wheel radius;

[0035] The calculation equation for acceleration is:

[0036]

[0037] In the formula, a is the acceleration of the car, v is the speed of the car, and t is the time;

[0038] Since the data signals collected by the sensor inevitably have errors, this will cause a large deviation between the calculated results and the actual results. Considering that the mass of the car is a slow variable, according to the relatively stable characteristics of the car mass change, the variables required in the mass estimation process are substituted into the recursive least squares system identification method equation, and the recursive least squares model of mass estimation is obtained:

[0039]

[0040] In the formula, is the estimated value of the vehicle mass at the kth moment, is the estimated value of the vehicle mass at the k-1th moment, y(k) is the system output at the kth moment, x(k) is the observable data vector, I is the unit matrix, K(k) is the gain matrix at the kth moment, and P(k) is the covariance matrix at the kth moment;

[0041] In order to improve the real-time quality estimation and prevent the occurrence of "data saturation", it is necessary to select a suitable forgetting factor. The commonly used empirical formulas of forgetting factors are:

[0042] λ(t)=1-0.05*0.98 t (11)

[0043] Where, t is the simulation time;

[0044] Since the range of the forgetting factor is usually [0.95, 1), considering the vehicle mass characteristics, the forgetting factor of the mass estimation model is selected as 0.985;

[0045] The specific process of step S3 is as follows:

[0046] Based on the longitudinal force analysis results of the vehicle in the turning condition, the turning resistance term is introduced into the traditional mass estimator, and the full-condition mass estimation model is established, in which the observation matrix of the system is:

[0047]

[0048] Where f is the rolling resistance coefficient, θ is the slope angle, δ is the mass conversion factor, R is the turning radius, and L is ris the distance from the center of mass to the rear axle, L f is the distance from the center of mass to the front axle, L is the wheelbase, α is the center angle of the curve, δ f is the front wheel turning angle, α r is the rear wheel slip angle;

[0049] The output of the system is m is the parameter to be estimated, and the full-condition quality estimation model is as follows:

[0050]

[0051] In the formula, is the estimated value of the vehicle mass at the kth moment, is the estimated value of the vehicle mass at the k-1th moment, y(k) is the system output at the kth moment, I is the unit matrix, K(k) is the gain matrix at the kth moment, and P(k) is the covariance matrix at the kth moment.

[0052] Compared with the prior art, the advantages of the present invention are:

[0053] 1 This method takes into account the common straight-line and turning conditions in actual road driving, introduces the turning resistance term into the straight-line condition quality estimator, and establishes a recursive least squares method model for all-condition quality estimation including the forgetting factor.

[0054] 2 This method enhances the robustness of the estimation method in actual driving and expands its scope of application. Especially in cornering conditions, the estimation accuracy of the full-condition parameter estimator is higher than that of the traditional parameter estimator. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The present invention will be further described below with reference to the accompanying drawings and examples:

[0056] Figure 1 The present invention is a flow chart of a method for estimating the mass of a pure electric vehicle in all operating conditions taking into account cornering conditions.

[0057] Figure 2 It is a schematic diagram of longitudinal force analysis of a method for estimating the mass of a pure electric vehicle in all working conditions taking into account a turning condition according to the present invention;

[0058] Figure 3 A schematic diagram of a recursive least square method with a forgetting factor for estimating the mass of a pure electric vehicle in all working conditions taking into account a turning condition according to the present invention; DETAILED DESCRIPTION

[0059] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present invention. In addition, in the following description, the description of well-known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present invention.

[0060] The present invention will be further described below in conjunction with the accompanying drawings.

[0061] See also Figure 1 The present invention provides a method for estimating the mass of a pure electric vehicle in all working conditions considering a turning condition, comprising the following steps:

[0062] S1, based on the vehicle driving external force and vehicle driving balance equation, establish the longitudinal dynamics model of pure electric vehicle;

[0063] S2, based on the longitudinal dynamics model, a mass estimation algorithm of the recursive least square method with forgetting factor under straight-line working conditions is established;

[0064] S3, based on the longitudinal force analysis of the vehicle under turning conditions, the turning resistance term is introduced into the straight-line condition estimator to establish a full-condition mass estimation model;

[0065] See also Figure 2 , the longitudinal forces during the driving process of the car specifically include the following:

[0066] When a car accelerates on a slope, the resistance and the driving force of the car are balanced. The car driving equation is as follows:

[0067] F t =F f +F w +F i +F j (1)

[0068] In the formula, F t is the driving force, F f is the rolling resistance, F w is the air resistance, F i is the slope resistance, F j For acceleration resistance;

[0069] The longitudinal dynamic model of the car during driving is:

[0070]

[0071] In the formula, F t is the longitudinal driving force of the car, m is the mass of the car, ρ is the air density, C d is the air resistance coefficient, θ is the road slope, and f is the rolling resistance coefficient;

[0072] Accelerate the longitudinal dynamics model of the vehicle in the above formula Differentiating it gives:

[0073]

[0074] Considering that the road slope changes little, the influence of the road slope on the acceleration is small during the driving process of the car, so the result of the slope on the acceleration can be approximated to zero and can be ignored, that is:

[0075]

[0076] The driving torque of pure electric vehicles changes very frequently. When the driving torque of the motor changes frequently, the rate of change of the vehicle acceleration is higher, while the rate of change of the vehicle speed is much smaller than the rate of change of the acceleration. Therefore, the derivative of the rate of change of the acceleration with respect to the rate of change of the speed is approximately zero and can be ignored, that is:

[0077]

[0078] The signal collected by the sensor is filtered. Since the same filter is used, according to the filter principle, combined with equations (4, (5)), equation (3) can be transformed into:

[0079]

[0080] In the formula, F * x is the driving force signal, v * x is the acceleration;

[0081] Using Newton's second law to analyze the forces during the driving process of a car is:

[0082] F=ma (7)

[0083] In the formula, F is the resultant force on the car during driving, m is the mass of the car, and a is the acceleration of the car.

[0084] The only parameters that can be transmitted from the vehicle controller via CANape are the output shaft speed and the actual output torque. When estimating the vehicle mass, the effects of air resistance, rolling resistance and road slope are ignored. The resultant force and acceleration of the vehicle required by Newton's second law can be calculated from the actual output torque and output shaft speed of the drive motor through the vehicle dynamics equation. The equation for calculating the driving force is:

[0085]

[0086] In the formula, F tFor the driving force of the car, T tq is the actual output torque of the driving motor, i g is the main reduction ratio of the automatic transmission, i0 is the transmission ratio of the automatic transmission, η is the mechanical efficiency of the transmission system, and r is the wheel radius;

[0087] The calculation equation for acceleration is:

[0088]

[0089] In the formula, a is the acceleration of the car, v is the speed of the car, and t is the time;

[0090] Since the data signals collected by the sensor inevitably have errors, this will cause a large deviation between the calculated results and the actual results. Considering that the mass of the car is a slow variable, according to the relatively stable characteristics of the car mass change, the variables required in the mass estimation process are substituted into the recursive least squares system identification method equation, and the recursive least squares model of mass estimation is obtained:

[0091]

[0092] In the formula, is the estimated value of the vehicle mass at the kth moment, is the estimated value of the vehicle mass at the k-1th moment, y(k) is the system output at the kth moment, x(k) is the observable data vector, I is the unit matrix, K(k) is the gain matrix at the kth moment, and P(k) is the covariance matrix at the kth moment;

[0093] In order to improve the real-time quality estimation and prevent the occurrence of "data saturation", it is necessary to select a suitable forgetting factor. The commonly used empirical formulas of forgetting factors are:

[0094] λ(t)=1-0.05*0.98 t (11)

[0095] Where, t is the simulation time;

[0096] Since the range of the forgetting factor is usually [0.95, 1), considering the vehicle mass characteristics, the forgetting factor of the mass estimation model is selected as 0.985;

[0097] See also Figure 3 , the recursive least squares method with forgetting factor under all working conditions includes the following:

[0098] Based on the longitudinal force analysis results of the vehicle in the turning condition, the turning resistance term is introduced into the traditional mass estimator, and the full-condition mass estimation model is established, in which the observation matrix of the system is:

[0099]

[0100] Where f is the rolling resistance coefficient, θ is the slope angle, δ is the mass conversion factor, R is the turning radius, and L is r is the distance from the center of mass to the rear axle, L f is the distance from the center of mass to the front axle, L is the wheelbase, α is the center angle of the curve, δ f is the front wheel turning angle, α r is the rear wheel slip angle:

[0101] The output of the system is m is the parameter to be estimated, and the full-condition quality estimation model is as follows:

[0102]

[0103] In the formula, is the estimated value of the vehicle mass at the kth moment, is the estimated value of the vehicle mass at the k-1th moment, y(k) is the system output at the kth moment, I is the unit matrix, K(k) is the gain matrix at the kth moment, and P(k) is the covariance matrix at the kth moment.

Claims

1. A method for estimating the mass of a pure electric vehicle in all working conditions considering a turning condition, characterized in that: The following steps are involved: S1, based on the vehicle driving external force and vehicle driving balance equation, establish the longitudinal dynamics model of pure electric vehicle; S2, based on the longitudinal dynamics model, a mass estimation algorithm of the recursive least square method with forgetting factor under straight-line working conditions is established; S3, based on the longitudinal force analysis of the vehicle under turning conditions, a turning resistance term is introduced into the straight-line mass estimator to establish a full-condition mass estimation model.

2. A method for estimating the full-operating mass of a pure electric vehicle considering a turning condition according to claim 1, characterized in that: The specific process of step S1 is as follows: When a car accelerates on a slope, the resistance and the driving force of the car are balanced. The car driving equation is as follows: F t =F f +F w +F i +F j (1) In the formula, F t is the driving force, F f is the rolling resistance, F w is the air resistance, F i is the slope resistance, F j For acceleration resistance; The longitudinal dynamic model of the car during driving is: In the formula, F t is the longitudinal driving force of the car, m is the mass of the car, ρ is the air density, C d is the air resistance coefficient, θ is the road slope, and f is the rolling resistance coefficient; Accelerate the longitudinal dynamics model of the vehicle in the above formula Differentiating it gives: Considering that the road slope changes little, the influence of the road slope on the acceleration is small during the driving process of the car, so the result of the slope on the acceleration can be approximated to zero and can be ignored, that is: The driving torque of a pure electric vehicle changes very frequently. When the driving torque of the motor changes frequently, the rate of change of the vehicle acceleration is higher, while the rate of change of the vehicle speed is much smaller than the rate of change of the acceleration. Therefore, the derivative of the rate of change of the acceleration with respect to the rate of change of the speed is approximately zero and can be ignored, that is, The signal collected by the sensor is filtered. Since the same filter is used, according to the filter principle, combined with equations (4, (5)), equation (3) can be transformed into: In the formula, F * x is the driving force signal, v * x is the acceleration.

3. The method for estimating the full-operating mass of a pure electric vehicle considering a turning condition according to claim 1, characterized in that: The specific process of step S2 is as follows: Using Newton's second law to analyze the forces during the driving process of a car is: F=ma (7) In the formula, F is the resultant force on the car during driving, m is the mass of the car, and a is the acceleration of the car; The only parameters that can be transmitted from the vehicle controller via CANape are the output shaft speed and the actual output torque. When estimating the vehicle mass, the effects of air resistance, rolling resistance and road slope are ignored. The resultant force and acceleration of the vehicle required by Newton's second law can be calculated from the actual output torque and output shaft speed of the drive motor through the vehicle dynamics equation. The equation for calculating the driving force is: In the formula, F t For the driving force of the car, T tq is the actual output torque of the driving motor, i g is the main reduction ratio of the automatic transmission, i0 is the transmission ratio of the automatic transmission, η is the mechanical efficiency of the transmission system, and r is the calculation equation for the wheel radius acceleration: In the formula, a is the acceleration of the car, v is the speed of the car, and t is the time; Since the data signals collected by the sensor inevitably have errors, this will cause a large deviation between the calculated results and the actual results. Considering that the mass of the car is a slow variable, according to the relatively stable characteristics of the car mass change, the variables required in the mass estimation process are substituted into the recursive least squares system identification method equation, and the recursive least squares model of mass estimation is obtained: In the formula, is the estimated value of the vehicle mass at the kth moment, is the estimated value of the vehicle mass at the k-1th moment, y(k) is the system output at the kth moment, x(k) is the observable data vector, I is the unit matrix, K(k) is the gain matrix at the kth moment, and P(k) is the covariance matrix at the kth moment; In order to improve the real-time quality estimation and prevent the occurrence of "data saturation", it is necessary to select a suitable forgetting factor. The commonly used empirical formulas of forgetting factors are: λ(t)=1-0.05*0.98 t (11) Where, t is the simulation time; Since the range of the forgetting factor is usually [0.95, 1), considering the mass characteristics of the vehicle, the forgetting factor of the mass estimation model is selected as 0.

985.

4. The method for estimating the mass of a pure electric vehicle in all working conditions considering a turning condition according to claim 1, characterized in that: The specific process of step S3 is as follows: Based on the longitudinal force analysis results of the vehicle in the turning condition, the turning resistance term is introduced into the traditional mass estimator, and the full-condition mass estimation model is established, in which the observation matrix of the system is: Where f is the rolling resistance coefficient, θ is the slope angle, δ is the mass conversion factor, R is the turning radius, and L is r is the distance from the center of mass to the rear axle, L f is the distance from the center of mass to the front axle, L is the wheelbase, α is the center angle of the curve, δ f is the front wheel turning angle, α r is the rear wheel slip angle; The output of the system is m is the parameter to be estimated, and the full-condition quality estimation model is as follows: In the formula, is the estimated value of the vehicle mass at the kth moment, is the estimated value of the vehicle mass at the k-1th moment, y(k) is the system output at the kth moment, I is the unit matrix, K(k) is the gain matrix at the kth moment, and P(k) is the covariance matrix at the kth moment.

Images